Monitoring of some major volatile organic compounds on board of

MONITORING OF SOME MAJOR VOLATILEORGANIC COMPOUNDS ON BOARD OFCHEMICAL TANKERS

W. Jacobs1,4, D. Dubois1,5, D. Aerts1,6, P. Declerck1,7, M. Stranger2,8, A. Buczynska2,9, A. Godoi3 and R. Van Grieken2,10Received 06 January 2010; in revised form 11 January 2010; accepted 04 June 2010

ABSTRACT

Nowadays, chemical tankers are transporting a wide variety of chemical prod-ucts. These products have different characteristics, and some are toxic. In litera-ture several studies suggest an increased incidence of various forms of cancer forcrew members on tankers. Most of these studies are based on medical statistics,rather than on effective measurements on board. The aim of this study is to moni-tor the average concentrations of toxic vapours on board of chemical tankers.Therefore we went on board of two chemical tankers for a 14-day trip and per-formed measurements during the normal operation of the vessel, in order to findout whether cargo vapours are traceable in the atmosphere in and around the ves-sel. The concentrations measured clearly show that there is a relationshipbetween the cargoes transported and the concentration of the cargo vapour in theatmosphere in and around the vessel. The results further show an elevated con-centration of the analyzed substances in the engine room and an important influ-ence of the relative wind direction. Tank vents situated in the gas-safe zone influ-enced the results as well.

Keywords: shipboard conditions, diffusive monitoring, occupational health,BTEX.

JOURNAL OF MARITIME RESEARCH 3

Journal of Maritime Research, Vol. VII. No. 2, pp. 3-20, 2010Copyright © 2010. SEECMAR

Printed in Santander (Spain). All rights reservedISSN: 1697-4840

1 Antwerp Maritime Academy, Noordkasteel-Oost 6, B-02030 Antwerp, Belgium. 2 Department of Chemistry, Micro andTrace Analysis Center, University of Antwerp, Universiteitsplein 1, B-2610 Antwerp, Belgium. 3 Professor, Federal University ofParana ([email protected]), Rua Francisco H. dos Santos, Curitiba, Brazil. 4 Captain,([email protected]). 5 Professor([email protected]). 6 Lecturer, ([email protected]) 7 Officer ([email protected]). 8 Doctor ([email protected]). 9 PhD student ([email protected]). 10 Professor ([email protected]).

1 agosto 2010:Maquetación 1 20/10/2010 20:29 Página 3

INTRODUCTION

Numerous gas and chemical tankers transport a wide variety of chemicals to feed theeconomy. It is inevitable that during common operations toxic cargo vapours arereleased into the atmosphere. But no recent data on their concentration in theatmosphere in and around ships are available.

In the literature there are several studies suggesting a causal relation between sail-ing as a crew member on tankers and the incidence of various forms of cancer (e.g.Nilsson et al., 1998; Saarni et al., 2002). Most of these studies are based on medicalstatistics. Only a few researchers went effectively on board to perform sampling andto identify the possible cause. The study of Moen et al. (1995a) was based on measure-ments of benzene performed on board. One can only hope that the high concentra-tions measured at that time have generally decreased thanks to improved operationprocedures. Examples are closed gauging, loading with high velocity valves or, evenbetter, with the use of a vapour return. In the case of high velocity valves, however, thereleased vapours are still present in the atmosphere nearby the ship. Our interest iswhether we could find any traces of these vapours back on board. Especially the studypublished in the International Safety Guide for Oil Terminals and Tankers, ISGOTT(ICS, OCIMF and IAPH (2006)) concerning air dispersion, may create some doubts.This study shows the air circulation around the ship’s superstructure. It might be pos-sible that parts of the cargo vapours released ahead of the superstructure are trans-ported behind it by natural air circulation. The inlets of the ventilation for the accom-modation (AC) and the engine room (ER) are situated at the aft end of the super-structure. One can assume that parts of these toxic vapours are being sucked by theventilation intake. Also the flue gasses from the ship’s engines as well as the vents ofthe different bunker tanks can be considered as a possible source of toxic vapours. Totest this assumption was the main drive for our investigation.

Another item that can be questioned is whether the segregation between thegas-dangerous (gdz) and the gas-safe zone (gsz) is as well defined as mentioned inthe international bulk chemical code (IBC)(IMO, (2007)). This subdivision is illus-trated in Figure 1.

The aim of this study is to give a general overview of the concentrations of thetoxic vapours at different locations on board, in various situations. In a later cam-

MONITORING OF SOME MAJOR VOLATILE ORGANIC COMPOUNDS ON BOARD ...

Volume VII. Number 2. year 20104

Figure 1: Gas-dangero s and gas-safe zones according to the IBC code


paign, we intend to study more in detail the concentrations in some specific spaces orzones that have been found interesting, based on the results of this campaign.

METHODS

Sampling and analytical methods

Diffusive sampling is particularly suited to determine time-weighted average volatileorganic compound (VOC) concentration in occupational hygiene and environmen-tal air monitoring (Oury et al., 2006; Pennequin-Cardinal et al., 2005). Moreover ithas as main advantages its simplicity and its cost effectiveness. Traditional diffusivesampling is characterized by low sampling rates and requires long sampling times.However, Radiello diffusive samplers are characterized by high and constant sam-pling rate values, high sensitivity and relatively short exposure times (Radiello UsersManual). The sampling rate is invariant towards humidity in the range 15-90% andwind speeds between 0.1 and 10 m.s-1 (Radiello Users Manual). This method is alsosuitable for multipoint and simultaneous measurements. The used Radiello diffusivesamplers consist of an adsorbing cartridge (530 mg of activated charcoal) inserted ina microporous polyethylene membrane. The diffusive sampler is exposed to air for ameasured time period. The rate of sampling for a specific compound is determinedby prior exposure in a standard atmosphere. VOCs migrate into the sampler by dif-fusion and are collected on the activated charcoal. In the laboratory, the collectedmaterial is desorbed by carbon disulfide. The solution is analyzed by gas chromatog-raphy coupled to mass spectrometry (GC-MS), according to a previously developedmethod ( Joos et al. 2003; Stranger, 2005).

A validation of this technique has been organized previously by our researchgroup. Here all samples were collected in triplicate, with an extra annular denudersystem nearby. The coefficients of variance, all having values below 10%, indicate thesufficient reproducibility for the collection of benzene, toluene, ethylbenzene andxylenes (BTEX) by means of Radiello passive samplers. The highest coefficients ofvariance, 12.2%, were calculated for the m+p-xylene determinations. However, allaverage BTEX concentrations agreed very well with the fixed monitor’s results, asthe ratios Radiello/denuder system were all close to one (Stranger, 2005).

Also three blank cartridges were carried along with each vessel. They were keptsealed and stored near the cartridges that have been used. Later analysis showed thatthe readings of these blank samplers were all below the detection limits and so nocorrection had to be made on the obtained results.

To monitor all of the VOC compounds in air quantitatively is very time consum-ing and expensive. We have chosen 8 aromatic hydrocarbons based upon their pres-ence in gasoline. As most cargoes are related to gasoline or are distillates from gaso-line, and since gasoline is also used in the engine, we expected to find higher concen-trations of these compounds on board. The compounds we analyzed the cartridges for

W. JACOBS, D. DUBOIS, D. AERTS, P. DECLERCK, M. STRANGER, A. BUCZYNSKA, A. GODOI AND R. VAN GRIEKEN



were benzene, toluene, ethylbenzene, m+p-xylene, o-xylene, 1,3,5-trimethylbenzene(1,3,5-TMB) and 1,2,4- trimethylbenzene (1,2,4 -TMB). Acute (short-term) expo-sure to gasoline and its components benzene, toluene and xylenes has been associatedwith skin and sensory irritation, central nervous system (CNS) problems (tiredness,dizziness, headache, loss of coordination), effects on the respiratory system and eyeand nose irritation. On top of skin, sensory and CNS problems, prolonged exposureto these compounds can also affect the kidney, liver and blood systems (Agency forToxic Substances and Disease Registry, 2004, 2007; Calabrese and Kenyon, 1991). Theseeffects are also a justification for the choice of the 8 analyzed compounds.

The ships

We had access to two chemical tankers, sister ships, with a length of 112 m,equipped with 17 cargo tanks and an overall capacity of just less than 10,000 m³.Both ships were trading between Rotterdam and the Baltic Sea area. The samplingplaces, chosen after a visit to the ships, are shown in Figure 2. These are the shelteron the maindeck, the manifold, the facing, the ventilation intake of the engine room(ER), the ventilation intake of the accommodation (AC), two different levels in theER and two in the AC. In order to get an image as complete as possible of the con-centrations, we took 63 samples per ship.

PS = portside, looking from aft to forward the left side of the vessel. SB = starboardside, looking from aft to forward the right side of the vessel

We made a logbook and a diary to keep records of the location of the tubes, theprevailing wind, weather conditions, voyage data, cargo operations, products on board,exposure time, etc. The temperature was automatically logged every 15 minutes.



Figure 2: The location of the diffusive samplers on board.


Study protocol

We changed the Radiello® adsorbing cartridges every time the pilot came on board onarrival in port, and every time the pilot disembarked on departure from a port. Thisway of working resulted in separate measurements for the time in port and the time atsea. We expected the results for the time spent at sea to be different from the results forthe time spent in port. Possible reasons for this difference might be the use of the mainengine, influence from pollution sources ashore, influences from other vessels and theinfluence of cargo related operations. For vessel A we have data from 2 sea voyages and3 ports, for vessel B data are available from 3 sea voyages and 3 ports.

For further classification of the results we divided a ship into four different areas.The first subdivision was in an indoor and outdoor area. Secondly indoor was split intoAC and ER, outdoor in a gdz and a gsz. Indoor space was split into AC and ER, as thefirst one has to be considered as a living environment, the second one as a workplace.Outdoor was subdivided as specified in the IBC code, the international code for theconstruction and equipment of ships carrying dangerous chemicals in bulk (IMO,2007). The gdz, being the area from forward to 3-5 m aft of the end of the superstruc-ture facing the cargo area, and the gsz being the area from aft to 3-5 m aft of the end of



Table 1: Cargo handling during the measurement period.

(a): hydrocarbon aliphatic fluid, 16-20 wt % aromatic content, 41 mg.kg-1 benzene content(b): a complex mixture of aliphatic, cycloparaffinic hydrocarbons (c): a highly refined lubricating mineral base oil, hydrocarbon fluid, with 34% weight 1,2,4-TMB(d): heavy alkanes C14-C17(e): nonylphenol + 9 EO polyethoxylate C9H19C6H4-O-(OCH2CH2)n-H (n=9)(f ): nonylphenol + 10 EO polyethoxylate C9H19C6H4-O-(OCH2CH2)n-H (n=10)

Ship Cargo Tanks Loading port to

Ship A toluene 8PS Antwerp to Raumaxylene 7PS Rotterdam to Raumaethanol 2PS, 4SB, 5PS, 5SB Rotterdam to Raumaisopropyl alcohol IPA 6PS Antwerp to Raumabutyl acrylate 3PS, 7SB Rotterdam to Haminamonopropylene glycol 2SB Rotterdam to Raumavarsol 40 (a) 3SB Antwerp to Raumaexxsol D60 (b) 4PS Antwerp to Raumasolvesso 150 (c) 6SB Antwerp to Raumasolvesso 100 (c) 8SB Antwerp to Rauma

Ship B phenol 4PS, 4SB Kotka to Rotterdamisoprene 1, 2PS Kotka to Rotterdamparaffin C14-C17 (d) 3SB, 6SB, 7PS, 8PS, 8SB Kotka to RotterdamNPE-9 (e) 7SB Kotka to RotterdamNPE-10 (f ) 5PS, 5SB Kotka to RotterdamNaOH 2PS, 2SB, 4PS, 4SB, 5PS Wilhelmshaven to Kotka


The benzene concentrations for ship A were all found to be below the detectionlimit.

the superstructure facing the cargo area, are shown in Figure 1. In order to reduce thenumber of results, we often used average values for one zone or space.

Table 1 gives an overview of the different products carried by the two vessels duringour campaign, with their respective stowage on board and loading and discharging port.

RESULTS

Port

In order to make clear which port has been called by which vessel, we will indicate(A) or (B) after each port and after each seavoyage. The results of the measurementsin port are shown in Table 2.



Table 2: Concentrations in port in µg.m-3.{exposure time in hours}.

( – = below the detection limit)

Port of Port of Port of Port of Port of Port ofIN PORT Zone Wilhelmshaven Kotka (B) Rotterdam Rotterdam Rauma (A) Hamina (A)

(B) {15 h} {108 h} (B) {11 h} (A) {24 h} {33 h} {52 h}benzene AC 5.01 2.06 62.1 – – –

ER 3.10 16.5 75.4 – – –Gsz 6.88 1.92 76.4 – – –Gdz 2.45 1.50 58.9 – – –

toluene AC 16.1 7.49 14.2 99.3 75.6 –ER 13.4 55.1 20.9 277 69.1 25.6Gsz 22.4 6.99 14.7 101 78.7 –Gdz 39.0 159 86.2 935 80.3 –

m+p-xylene AC 15.8 7.37 10.5 85.7 70.0 18.6ER 15.2 76.0 22.7 324 1550 78.7Gsz 24.5 7.59 10.8 104 56.8 15.3Gdz 6.46 46.2 2.93 932 73.2 –

o-xylene AC 6.28 1.87 4.41 55.4 23.8 –ER 5.86 23.6 9.56 275 314 13.6Gsz 9.63 1.73 4.58 57.8 18.5 –Gdz 2.58 1.13 1.53 418 25.4 –

ethylbenzene AC 12.2 2.43 7.55 17.9 23.5 –ER 11.3 51.8 17.2 173 309 22.7Gsz 19.1 2.41 9.41 38.3 20.6 –Gdz 5.10 1.19 2.54 811 31.6 –

1,3,5-TMB AC 3.98 0.857 2.55 – – –ER 3.69 16.0 6.12 68.5 45.6 20.0Gsz 6.22 0.736 2.41 12.8 – –Gdz 1.13 0.247 0.376 270 6.86 –

1,2,4-TMB AC 12.6 2.89 8.41 77.7 38.5 18.9ER 12.2 52.0 20.0 435 293 123Gsz 19.6 2.45 7.10 47.8 21.9 9.51Gdz 2.90 0.767 1.11 1794 53.0 –




( – = below the detection limit)

Sea voyage

The results of the measurements at sea are displayed in Table 3. The benzene con-centrations for ship A were all found below the detection limit except for two meas-urements in the AC at sea, location AC 0. Although some of the transported prod-ucts might contain a small amount of benzene, its vapour is not detectable in ourmeasurements.

Table 3: Concentrations at sea in µg.m-3 {exposure time in hours}.

The results for benzene on ship B reflect a concentration below or near to thedetection limit. Two exceptions are the ER during the voyage from Wilhelmshavento Kotka (B) and from Kotka to Rotterdam (B).

DISCUSSION

During normal cargo operations some cargo vapours are released into theatmosphere. These vapours should be traceable in the gdz. What we wanted to

Seavoyage Seavoyage Seavoyage Seavoyage SeavoyageRotterdam Wilhelmshaven Kotka Rotterdam Rauma

AT SEA Zone Wilhelmshaven Kotka Rotterdam Rauma Hamina(B) {19h} (B) {76h} (B) {93h} (A) {93h} (A) {33h}

Benzene AC 2.21 3.27 2.10 3.27 1.65ER 2.52 22.1 9.72 – –Gsz 1.59 1.94 1.49 – –Gdz – 1.18 1.26 – –

toluene AC 7.75 7.85 4.71 16.9 291ER 10.3 41.7 33.0 11.3 608Gsz 6.81 6.80 6.31 13.5 471Gdz 74.4 338 343 58.4 385

m+p-xylene AC 6.49 12.7 3.56 44.8 86.9ER 9.99 38.1 40.2 109 261Gsz 5.92 4.34 3.88 39.5 141Gdz 3.11 7.80 6.26 531 157

o-xylene AC 2.62 7.70 2.31 4.36 1.30ER 3.77 11.9 13.3 23.9 49.8Gsz 2.18 1.64 1.39 12.2 34.5Gdz 1.28 2.34 1.43 178 32.9

ethylbenzene AC 3.47 5.83 1.88 7.51 22.8ER 6.78 33.4 23.4 22.5 56.7Gsz 3.29 2.92 2.29 15.4 27.3Gdz 0.211 2.65 1.99 191 43.2

1,3,5-TMB AC 1.05 2.84 0.970 3.55 15.3ER 2.15 6.56 9.18 11.5 37.9Gsz 0.944 1.02 0.926 3.55 16.7Gdz – 0.301 0.369 30.8 20.0

1,2,4-TMB AC 3.49 8.53 3.28 21.8 85.4ER 7.55 20.0 29.5 72.5 250Gsz 3.06 3.27 3.05 16.9 103


investigate is whether traces of these cargo vapours can also be found in other areason board of the vessel. The first zone that has been evaluated is the gsz.

A comparison between the gsz and the gdz

In order to compare both zones, the (gdz/gsz) ratio will be used. We will first discussthe results ship by ship, and draw up a conclusion later by comparing both results.

Ship A

Ship A transported cargoes whereof the vapours had a direct influence on our meas-urements, namely toluene, xylene, solvesso 100 (34 wt% 1,2,4-TMB). As a generalresult for this vessel, we found that the concentration in the gdz exceeded the con-centration in the gsz. Exceptions to this rule are the port of Hamina and the tolueneand o-xylene concentration between Rauma and Hamina. This is shown in Table 4.

Table 4: Concentration ratios for the gdz to the gsz for ship A



* indicates that the concentration in the gsz equals 0

From table 4 we learn that in Rotterdam and during the voyage from Rotterdamto Rauma the gdz/gsz ratio was situated well above unity, while in Rauma and dur-ing the voyage from Rauma to Hamina the same ratio was close to one, or evensmaller than one. For both ports, the relative wind direction was the same, abeamfrom starboard, and therefore this parameter cannot be responsible for the differ-ence. The major reason is the fact that during loading operations a considerableamount of cargo vapours are released on deck (gdz > gsz), while during dischargingoperations this is considerably less or even nil (gdz > gsz). When loading, the ship’stank is filled up, and consequently the pressure inside the tank will rise. At a presetpressure the relief valve will evacuate the tank vapours at a speed of 30 m.s-1 into theatmosphere of the gdz. Thanks to the true wind coming in from the starboard side,most of the vapours were evacuated to the portside, and not into the gsz. This result-ed in Rotterdam in an important difference in concentrations between the gdz andthe gsz. When discharging, the ship’s tank is emptied, and a negative pressure is cre-ated inside the tank. The vacuum valve will at a preset negative pressure open, andallow air into the tank to partially fill up the vacuum. During this operation no cargo

Ratio gdz/gsz Benzene Toluene m+p- o-Xylene Ethyl - 1,3,5- 1,2,4-Xylene benzene TMB TMB

Port of R’dam (A) * 9.2 9.0 7.2 21 21 37Sea voyage R’dam-Rauma (A) * 4.3 13 15 12 8.6 13Port of Rauma (A) * 1.0 1.3 1.4 1.5 * 2.4Sea voyage Rauma-Hamina (A) * 0.8 1.1 0.9 1.5 1.1 1.2Port of Hamina (A) * * 0.0 * * * 0.0


vapours will escape from the tank and therefore the influence of the cargo on ourmeasurements in Rauma is rather small.

Comparing both sea voyages is more difficult. The voyage from Rotterdam toRauma was a laden voyage. Due to ship movements small amounts of toluene, xyleneand 1,2,4-TMB vapours were released in the gdz. We noticed that only fractions fromthese concentrations were measured in the other zones. This is illustrated in Figure 3.

The voyage from Rauma to Hamina was a voyage with extensive tank cleaningand tank ventilation. Considerable amounts of toluene, xylene and 1,2,4-TMBvapours have been ventilated into the gdz. We expected here also a greater value inthe gdz compared to the gsz. But the results in Figure 4 clearly show that the con-centration in both zones is almost equal or even a little bit higher in the gsz. As wedefinitely know that the main sources here are the tanks, we conclude that the venti-lated cargo vapours entered the gsz, and even show a tendency to accumulate there.Accordingly the concentration in the 3 other zones is remarkably higher than duringthe previous voyage.

Another exception is the port of Hamina. As can be found in Table 2, all valuesfrom both zones are close to the detection limit, reducing the significance of thegdz/gsz ratio. The reason why the concentrations here are situated near the detectionlimit can be explained by the fact that all cargoes with a direct influence on the

measurements have been dis-charged and cleaned earlierand the fact that there is onlylittle industrial activity inHamina.

Ship B

For ship B none of the trans-ported cargoes did have adirect influence on the meas-urements. While on ship A ingeneral gdz concentrationsexceeded the gsz ones, wefound on ship B the opposite.Exceptions were the tolueneand the m+p-xylene concen-trations in Kotka. Theseresults are shown in Table 5.Also for the o-xylene concen-trations for Wilhelmshaven toKotka and for Kotka to Rot-terdam, the gdz value exceed



Figure 3: Concentrations in µg.m-3 for the different zones fromRotterdam to Rauma, ship A.

Figure 4: Concentrations in µg.mm-3 for the different zones fromRauma to Hamina, ship A.


those of the gsz. But since the absolute concentrations are very low, varying from1.39 to 2.34 µg.m-3 (see Table 2), we found this less relevant.

Table 5: Concentration ratios for the gdz to the gsz for ship B

The ratio clearly shows a toluene pollution source situated in the gdz. Unlikeship A, no toluene cargo was on board of this ship. However we found an intermedi-ate bulk container (IBC) on deck, containing approximately 1 m³ of toluene. Thistoluene is sometimes used for tank cleaning purposes. An inspection of this IBCtaught us that the container was not gastight closed as the cap was broken. In thisway toluene vapours were released continuously, explaining the relatively high con-centrations for toluene in the gdz. The IBC was situated within 10 meter distancefrom our sampler. This situation looks very similar to the laden voyage of ship A,where only the sampler near the facing showed an increased concentration for theproducts transported. The higher m+p-xylene ratio in Kotka was due to an increasedgdz value. An explanation for this increment has not been found.

In contrast with ship A, the gdz/gsz ratio on ship B is in general below 1. Andalthough the absolute concentrations for ship B are lower, it is a contradiction thatthe gdz shows lower concentrations than the gsz. This was definitely not the inten-tion of the IBC code. The ratio indicates that the major source is closer to or situatedin the gsz.

Evaluation of the ER

The results found on ship B in the gsz were expected to be lower compared tothe gdz. However, the concentrations in the ER of ship B are the highest of all fourzones on the exception of toluene in the gdz. May we consider the ER as the majorvoc source on board of ship B? In order to confirm this statement, we calculated theindoor/outdoor (I/O) ratio for the ER. This ratio is shown in Table 6.

In most cases the ratio is well above 1 in spite of the powerful mechanical venti-lation present in the ER. The I/O ratio is in most cases above unit, indicating thatthe engine room itself is the major source and the ventilation here is insufficient orinefficient. The I/O ratio is only in two locations below unit, namely in Wil-helmshaven and for benzene in Rotterdam. The results in Wilhelmshaven are appar-



Ratio gdz/gsz Benzene Toluene m+p- o-Xylene Ethyl - 1,3,5- 1,2,4-Xylene benzene TMB TMB

Seavoyage R’dam-W’haven(B) 0.0 10.9 0.5 0.6 0.1 0.0 0.1Port of W’haven(B) 0.4 1.7 0.3 0.3 0.3 0.2 0.1Seavoyage W’haven-Kotka(B) 0.6 49.7 1.8 1.4 0.9 0.3 0.3Port of Kotka(B) 0.8 22.8 6.1 0.7 0.5 0.3 0.3Seavoyage Kotka-R’dam(B) 0.8 54.3 1.6 1.0 0.9 0.4 0.4Port of R’dam(B) 0.8 5.8 0.3 0.3 0.3 0.2 0.2


ently influenced by pollution from outside since the main engine is not running. Thetoluene/benzene ratio in Table 7 confirms that both compounds are coming fromthe same source, except of course for the facing where the sampler has been influ-enced by the small amounts of toluene escaping from the IBC. The presence of thecoal driven power plant nearby might be the main cause for this outdoor pollution.

In Rotterdam we measured unex-pectedly high benzene concentrations atall samplers, indicating that the wholeship was “covered” by a same level ofbenzene vapours. As there was no ben-zene cargo on board, and as never before

such high concentrations of benzene have been measured on board, it seems obviousthat this pollution came from shore.

Besides these two cases where the atmosphere around the vessel has been underinfluence of shore pollution, we find on ship B an I/O ratio for the ER (well) above1. Regretfully we could not make the same comparison for ship A as the results forinlet ER ship A were disturbed by an excessive air flow from the ER ventilation.

Based on the results of ship B we conclude that in the ER important pollutionsources are present. Further detailed research in the ER is necessary in order to iden-tify these sources.

Positioning of the ventilation inlets

As we measured quite a difference in absolute concentrations between the ventila-tion inlet of the AC and the ventilation inlet of the ER, we tried to determine whichof both inlets was located best. Both concentrations are only available for ship B, asthe concentrations near the inlet ER on ship A were rejected due to a too high venti-lation speed. The results for ship B are shown in Table 8.

Both ventilation intakes are somewhat 20 meters apart. But more important isthe fact that the AC intake is situated about 8 meters higher than the ER intake.The latter is situated 4 m more aft and on the starboard side. This is the side wherewe placed our sampler and it is the intake which is most frequently used. The AC



Table 6: Indoor/outdoor ratio in the ER for ship B.

Table 7: Toluene - benzene ratio in Wilhelmshaven

I/O ER Benzene Toluene Ethyl - m+p- o-Xylene 1,3,5- 1,2,4-benzene Xylene TMB TMB

R’dam/W’haven 1.6 1.6 2.2 2.0 2.0 2.7 3.1W’haven 0.3 0.5 0.4 0.5 0.5 0.5 0.5W’haven/Kotka 11.3 4.9 11 7.4 6.3 6.3 6.0Kotka 11 9.0 50 16 19 467 42Kotka/R’dam 4.8 3.1 6.1 6.3 5.9 5.8 5.7R’dam 0.8 2.3 2.7 3.6 3.2 4.9 5.3

AC 0 AC bridge inlet AC inlet ER Facing

3.3 3.1 3.2 3.2 16


intake is situated on the portside. From Table 8 we learn that there is a direct relationbetween the ratio inlet AC/inlet ER and the relative wind direction. When this rela-tive wind is forward of the beam or ahead, we noticed that the ratio is smaller than 1.If on the other hand, the relative wind is coming in from a direction abaft the beamor from astern, the ratio is found to be above 1. We have seen before that the influ-ence of the cargo on our measurements on board of this vessel is rather small. Theexhaust gases from the main engine and the auxiliaries together with possible shorepollution in port can be considered as the major sources. This is in agreement withthe relative position of both inlets compared to the position of the funnel. In termsof percentage, a wind from forward of the beam has a higher frequency of occur-rence than a wind from abaft the beam, due to the ship’s speed. Therefore we mayconclude that the position of the AC inlet is a better choice, although not ideal in allcircumstances. One exception on the above described relation is the benzene con-centration in Rotterdam. This is a confirmation of a previous conclusion, namelythat in Rotterdam a shore benzene pollution influenced our measurements. Theconcentrations near the AC inlet were clearly lower than near the ER room inlet. Inview of the difference in height, the relative vapour density of benzene (2.7) is a pos-sible explanation. So the result in Wilhelmshaven where also shore pollution wasconsidered could be explained by this same reason.

Evaluation of the AC

An interesting space to discuss more in detail is the AC. The result of the samplersplaced inside the AC, one at level 0, and one at the bridge level, should give ananswer to the question whether cargo vapours are traceable in the ER.

As shown in Table 9, there is a good correlation between the results of level 0and the bridge level concentrations on ship A. This correlation is somewhat lowerfor m+p-xylene, although both locations have the same ventilation inlet. Moreoverthe absolute concentrations at level 0 were always found to be greater than thebridge level concentrations. Opening and closing of doors for deck access at level 1 isa possibility for an increased concentration.



Table 8: Ratio inlet AC/inlet ER for the different compounds and different measuring periods with the corresponding wind direction and wind force on board of ship B.

Position Benzene Toluene Ethyl- m+p- o- 1,3,5- 1,2,4- Rel. wind Rel. windbenzene Xylene Xylene TMB TMB direction force

Kotka 1.5 1.3 3.7 2.2 1.7 3.2 3.0 110 SB 2R’dam 0.7 2.3 1.9 2.5 2.1 2.9 2.8 160 PS 3W’haven 0.5 0.5 0.5 0.6 0.6 0.6 0.6 60 SB 3Kotka/ R’dam 0.5 0.2 0.2 0.2 0.2 0.2 0.2 25 PS 4R’dam/W’haven 1.0 1.1 1.1 1.4 1.3 1.4 1.5 155 PS 4W’haven/Kotka 1.0 0.6 0.9 0.7 0.7 0.9 1.0 20 SB 3


The presence of cargo vapours in the AC is visible in the indoor/outdoor ratios ofTable 10. As expected, the indoor/outdoor ratio in Rotterdam is below unit, due to thelocal pollution outside. However an exception is 1,2,4-TMB. But when we keep inmind that Solvesso 100 has been loaded in Antwerp only 14 hours ago, it is possiblethat remnants of this operation are still present inside the AC, resulting in a relativelyhigh ratio. Once at sea, we would expect lower concentrations outside and thus a ratioabove unit. This is particularly the case for the cargo related vapours, toluene, xyleneand 1,2,4-TMB. This is another indication that cargo vapours entered the AC duringthe stay in Rotterdam. In Rauma the concentrations at the AC inlet hardly differ fromthe inlet concentrations at sea, except for toluene. This means that the air quality inRauma is almost as good as at sea, resulting in an indoor/outdoor ratio equal to orabove unity. When we started cleaning and venting the tanks on the way to Hamina,the ratios decreased below one. As can be concluded from Table 10, particularly thevapours related to the transported cargoes show a lower ratio, proving that cargovapours arrived at the ventilation inlet of the AC . Finally we arrived in Hamina withall toluene, xylene and Solvesso 100 tanks cleaned and made gas free. And as theindustrial activity in Hamina is rather low, the ratios became again above unity, at leastfor these results where the measured concentrations were above the detection limit.

The results of the bridge sampler and the level 0 sampler on ship B are stronglycorrelated as shown in Table 11. Both have the same ventilation intake, and we donot expect different sources for both spaces. The indoor/outdoor ratio shown inTable 12 teaches us that in general on ship B this ratio is below unity in port, aboveunity at sea. This is an understandable result, as we may consider the air at sea to beless polluted compared to the air in port. This statement is somewhat less evident forthe benzene concentration indicating the presence of a benzene source or accumula-tion inside. The relatively high percentage of smoking crewmembers might beresponsible for that result.



Table 9: Pearson correlation between samplers at AC level 0 and bridge level on board of ship A.

*outdoor concentration equals 0

Table 10: Indoor/outdoor ratio for AC on board of ship A.

Toluene Ethylbenzene M+p-Xylene o- Xylene 1,3,5-TMB 1,2,4-TMB

0.995 0.950 0.821 0.973 0.991 0.984

I/O AC Benzene Toluene Ethyl - m+p- o-Xylene 1,3,5- 1,2,4-benzene Xylene TMB TMB

R’dam 1.0 0.9 0.8 1.0 0.0 1.6R’dam/Rauma 2.5 1.0 1.1 1.1 1.0 1.3Rauma 1.0 1.1 1.2 1.3 * 1.7Rauma/Hamina 0.6 0.8 0.6 0.6 0.9 0.8Hamina * * 1.2 * * 2.0


From the AC indoor/outdoor ratio on ship A and ship B we learned that in someparticular cases, cargo vapours do enter the AC, that generally spoken in port the con-centration indoor is below outdoor concentration, and that at sea it is just the otherway around. Particularly cleaning and venting of cargo tanks leads to increased outdoorconcentrations, and as a consequence also to increased indoor concentrations.The idealwould be if we could fully isolate the AC from all toxic vapours outside. This is ofcourse hypothetical as the crew needs access to the deck part. Naval architects havethought about a solution to have access to deck without vapours from outside enteringthe AC. This is called the ‘crew changing room’. It is a small compartment situated atlevel 1 on the starboard side, with a door on one side giving access to the AC, on theother side a door giving access to the main deck. This compartment acts as an air lock.According to the ventilation plan, this room has a ventilation capacity of 400m³.h-1

resulting in 27 air changes per hour. Another use of this compartment is that the crewchanges their workclothes used outside, with clothes for indoor use and leaves the used

clothes also in this compart-ment in order to bring themlater to the laundry. As it isrealistic that in this spacehigher concentrations couldoccur, we placed on ship B anextra sampler and the resultsare shown in Figure 5. Somepeak values can be found inthis figure. Especially the seavoyages Wilhelmshaven toKotka and Kotka to Rotter-



Table 11: Pearson correlation between the samplers at AC level 0 and at the bridge level on ship B.

Table 12: Indoor/outdoor ratio for AC on board of ship B.

Figure 5: Concentrations in the changing room of ship B.

I/O AC Benzene Toluene Ethyl - m+p- o-Xylene 1,3,5- 1,2,4-benzene Xylene TMB TMB

R’dam/W’haven 1.3 1.1 1.0 0.9 0.9 0.9 0.9W’haven 1.1 1.1 0.9 0.8 0.8 0.8 0.8W’haven/Kotka 1.7 1.5 2.0 3.6 3.1 2.9 2.7Kotka 0.9 0.9 0.6 0.7 0.8 0.8 0.8Kotka/R’dam 1.7 2.4 2.5 2.5 2.4 3.5 3.4R’dam 1.0 0.7 0.6 0.7 0.7 0.7 0.8

correlation AC-bridge Benzene Toluene Ethyl - m+p- o-Xylene 1,3,5- 1,2,4-benzene Xylene TMB TMB

1.000 0.978 0.991 0.974 0.993 0.981 0.957


dam draw our attention. The ethylbenzene, xylenes and 1,2,4-TMB concentrationswere significantly higher. In an attempt to find out whether all compounds are origi-nating from the same source, Table 13 shows ratios between the analyzed compoundsand respectively toluene and benzene. Toluene and benzene concentrations are relatedto another source, at least for the two above mentioned voyages.

Although the ventilation in the changing room is completely separated from theventilation in the AC, we will compare both concentrations. The sampler placed inthe AC on level 0 is located somewhat 8 meters away from the changing room. Anydifferences in concentrations between the two samplers are most probably in relationwith the quality of the supplied air. The correlation between both concentrationsover the whole measuring campaign is shown in Table 14. This correlation confirmsthe conclusion from Table 13. For benzene and toluene there is a good to even per-fect correlation.

For the other compounds there is no correlation at all, confirming that theseresults were influenced by a different source. When we compare the absolute concen-trations of the AC level 0 and the absolute concentration in the changing room forthose compounds that do not show any correlation, the changing room concentrationsare during 2 seavoyages definitely higher, namely from Wilhelmshaven to Kotka andfrom Kotka to Rotterdam. During these two voyages the ratio changing room overAC level 0 varies between 1.8 and 18. There is a source of ethylbenzene, xylenes andtrimethylbenzene present in the changing room. As none of these products were relat-ed to the cargo, the source must be different. A first possibility is cigarette smoke fromthe duty mess nearby. But a further study of the ventilation plan showed a very surpris-



Table 13: Ratios for the measured compounds to toluene and benzene in the changing room on ship B.

Table 14: Pearson correlation between the concentrations measured in the AC and in the changing room on ship B.

Compound ratios in changing room

R’dam/W’haven 0.2 0.3 0.9 0.3 0.1 0.4 1.5 4.6 1.3 0.6 2.0W’haven 0.3 0.7 1.0 0.4 0.2 0.8 2.4 3.3 1.3 0.8 2.6W’haven/Kotka 0.3 1.7 6.2 1.2 0.5 1.6 6.0 21.9 4.4 1.8 5.6Kotka 0.3 0.3 1.1 0.3 0.2 0.5 1.2 4.0 1.1 0.6 1.9Kotka/R’dam 0.2 2.5 7.4 1.7 3.1 9.9 12 37 8.4 15 49R’dam 3.7 0.6 0.7 0.3 0.1 0.4 0.2 0.2 0.1 0.0 0.1

Benz

ene/

tolu

ene

Ethy

l-ben

zene

/to

luen

e

m+p

-Xyl

ene/

tolu

ene

o-Xy

lene

/ tol

uene

1,3,

5-TM

B/to

luen

e

1,2,

4-TM

B/to

luen

e

Ethy

l-ben

zene

/be

nzen

e

m+p

-Xyl

ene/

benz

ene

o-Xy

lene

/ be

nzen

e

1,3,

5-TM

B/be

nzen

e

1,2,

4-TM

B/be

nzen

e

Correlation between the AC Benzene Toluene Ethyl - m+p- o-Xylene 1,3,5- 1,2,4-and changing room benzene Xylene TMB TMB

1.000 0.979 0.102 -0.090 -0.007 -0.246 -0.255


ing result. The intake of the supply of the chang-ing room is situated just above the vents of thepolluted water tank and thermal oil tank, asshown in Figure 6. The polluted water tank con-tains a mix of water and all kinds of other liquidsthat cannot be discharged directly because of eco-logical reasons. These liquids are temporarilystored in the polluted water tank for decantation.Examples of other liquids are bilges water or liq-uids from any leak in the ER. It is not uncommonthat these liquids contain hydrocarbons. In orderto accelerate the segregation/decantation withwater, this tank is kept at a temperature between50°C and 70°C. Therefore it is not surprising thatdifferent compounds do have a higher concentra-tion in the changing room. We would thereforeadvice the crew not to use the supply for thechanging room, but only the extraction fan. Our

results show that the air in the gdz is cleaner than the air near the polluted water tankvent. This error in the configuration of the ventilation raises some doubts whethernaval architects are really concerned about the seafarers’ health.

CONCLUSIONS

Comparing ship A with ship B shows that the cargo has a very important influenceon the absolute concentrations. During most of the cargo handling the influence ofcargo vapours is restricted to the gdz, but when loading and especially during clean-ing and ventilating the tanks, the cargo vapours can be found all over the vessel. Asecond important pollution source is the ER. Although the absolute concentrationsfound are well below these of the cargo, the results clearly show that there is a prob-lem in the ER despite the mechanical ventilation. A third pollution source, depend-ing on the location, is the shore. The loading and discharging operations mostly takeplace in an industrial area, and sometimes really close to the production or treatmentplant. The example given of the crew changing room ventilation arises the doubtwhether naval architects are sufficiently concerned about the seafarers’ health. Ourconclusion is that the actual ventilation configuration can be improved.

ACKNOWLEDGEMENTS

The authors would like to thank Ahlers Maritime Services Antwerp, for the hospi-tality and the cooperation on board. The authors also would like to thank the Flem-ish Institute for the Sea (VLIZ), for providing measuring apparatus.



Figure 6: Positioning of inlet changingroom relative to vent polluted watertank and thermal oil tank.


REFERENCES

Agency for Toxic Substances and Disease Registry (ATSDR). (2007). Toxicological profile forbenzene. Atlanta, Georgia: U.S. Department of Health and Human Services, PublicHealth Service.

Agency for Toxic Substances and Disease Registry (ATSDR). (2004). Interaction profile forbenzene, toluene, ethylbenzene and xylene (BTEX). Atlanta, Georgia: U.S. Department ofHealth and Human Services, Public Health Service.

Agency for Toxic Substances and Disease Registry (ATSDR). (2007). Toxicological profile forethylbenzene. Atlanta, Georgia: U.S. Department of Health and Human Services, PublicHealth Service.

Calabrese EJ, Kenyon EM. (1991) Air Toxics and Risk Assessment. Lewis Publishers, Chelsea,MI. ISBN 0873711653.

ICS, OCIMF, IAPH. (2006) ISGOTT 5th ed. International Safety Guide for Oil Tankers &Terminals. Witherby Marine Publishing, London.

IMO. (2007) International code for the construction and equipment of ships carrying dangerouschemicals in bulk (IBC Code). IMO publications, London.

Joos PE, Godoi AFL, De Jong R, de Zeeuw J, Van Grieken R. (2003) Trace analysis of ben-zene, toluene, ethylbenzene and xylene isomers in environmental samples by low-pres-sure gas chromatography–ion trap mass spectrometry. Journal of Chromatography A; 985,191-196.

Moen BE, Hollund BE, Berntsen M, Flo R, Kyvik KR, Riise T. (1995a). Exposure of thedeck crew to carcinogenic agents on oil product tankers. The Annals of OccupationalHygiene; 39 347-361.

Moen BE, Bjorg E, Hollund BE, Berntsen M, Flo R, KyvikK R, and Rilse T. (1995b). Occu-pational exposure of deck crews to carcinogenic agents on crude oil tankers. AmericanJournal of Industrial Medicine; 27 555-564.

Nilsson RI, Nordlinder R, Hörte LG, Järvholm B. (1998). Leukaemia, lymphoma, and multi-ple myeloma in seamen on tankers. Occupational Environment Medicine; 55 517-521.

Oury B, Lhuillier F, Protois J-C, Moréle Y. (2006). Behavior of the GABIE, 3M 3500,PerkinElmer Tenax TA, and Radiello 145 diffusive samplers exposed over a long time toa low concentration of VOCs. Journal of Occupational and Environmental Hygiene;10 547-57.




Pennequin-Cardinal A, Plaisance H, Locoge N, Ramalho O, Kirchner S, Galloo J.(2005).Performances of the Radiello diffusive sampler for BTEX measurements: Influence ofenvironmental conditions and determination of modelled sampling rates. AtmosphericEnvironment; 39 2535-2544.

Saarni H, Pentti J, Pukkala E. (2002). Cancer at sea: a case-control study among male Finnishseafarers. International Archives of Occupational and Environmental Health; 59 613-619

Stranger M. (2005). Characterisation of health related particulate and gasphase compounds inmultiple indoor and outdoor sites in Flanders. Thesis (PhD), University of Antwerp, Bel-gium.

Victorin K. (1993). Health effects of urban air pollutants guideline values and conditions inSweden. Chemosphere; 27 1691–1706.




Documents

Monitoring of some major volatile organic compounds on board of